Memory System Poison Flag Data Corruption Indication

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing memory systems face challenges in efficiently identifying and storing corrupted data without using metadata, which leads to high latency and reduced available storage space.

Innovation Solution

The memory system is configured to identify and store corrupted data by using indications such as poison flags, eliminating the need for metadata, thereby increasing storage space and reducing latency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If metadata is used to identify corrupted data, then data corruption identification is improved, but storage space is reduced and latency increases

Engineering Contradiction:
Improvedata corruption identificationVSAvoidstorage space
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts the corrupted data identification function from the metadata storage structure. Instead of storing metadata in the memory array, the system uses a separate indication mechanism (poison flag) that points to corrupted data without consuming storage space within the memory array, thereby resolving the contradiction between identification capability and storage space.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an intermediary indication structure (poison flag table) that mediates between the host system and memory array. This intermediary stores corruption information externally, allowing the memory array to maintain full storage capacity while still enabling corrupted data identification through the flag table's pointers.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If metadata is used to identify corrupted data, then data corruption identification is improved, but access latency increases

Engineering Contradiction:
Improvedata corruption identificationVSAvoidaccess latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-establishing the poison flag table structure before data storage operations. The indication mechanism is prepared in advance, allowing corrupted data to be quickly identified through simple pointer lookup rather than searching through metadata during data access, thereby reducing latency while maintaining identification capability.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If metadata is stored in the memory array, then data corruption identification is improved, but available storage space decreases

Engineering Contradiction:
Improvedata corruption identificationVSAvoidavailable storage space
Core Design Contradiction:
ReliabilityVSVolume of stationary object

Solution Approach 1:

The patent segments the storage system into two functional parts: the memory array for data storage and the poison flag table for corruption indication. This segmentation allows the memory array to dedicate 100% of its capacity to user data while the flag table externally manages corruption information, eliminating the trade-off between metadata storage and available space.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS20250117273A1Indicating data corruption
Publication Date: 2025.04.10 MICRON TECHNOLOGY INC
  • US20250117273A1 patent drawing
  • US20250117273A1 patent drawing
  • US20250117273A1 patent drawing

AI summary

Methods, systems, and devices for indicating data corruption are described. A memory system may be configured to identify and store corrupted data received from a host system without storing metadata. As part of transmitting a bulk transmission, the host system may transmit first data to be stored at an address of the memory system, and a first indication identifying that the first data is corrupted. The memory system may generate second data with a pattern of bits indicating that data stored at the address of the memory system is corrupted. The memory system may store the second data to the address, and later retrieve the second data in response to receiving a read command from the host system. Then, the memory system may generate a second indication identifying that the second data is corrupted, and transmit the second data and the second indication to the host system.